A Four-Acre Food Commons: Year-One Economics of a Polytarp-Tunnel, Hydroponic, and Poultry-Integrated Vegetable System
A preliminary, auditable estimate of how many families a mixed system can provision in its first year—before the orchard matures—using bed-area benchmarks, cycle throughput, and upper-bound stocking
Abstract
This post sets out a first-year economic and productive estimate for a four-acre property designed to support multiple families through a communal provisioning model. The production core is protected cropping in a 1,000 m² polytarp tunnel with 18 full 40 m² plots plus two 20 m² plots (760 m² productive bed area), supported by an additional 200 m² outdoor vegetable area (total 960 m²). Capacity is translated into “families fed” using The Diggers Club’s published rule-of-thumb that 10 m² of productive garden bed can feed one person and 40 m² can feed a family of four. On that basis, the system’s vegetable bed area corresponds to roughly 24 families (96 people) in the first year, before any orchard yield is counted. The tunnel also includes a hydroponic greens module (3 m × 3 m well-watered area) producing around 1,200 units of mixed greens per cycle under standard lettuce/rocket timings, plus vertical growing surfaces, and pest control via quail and bantam chickens. Alongside vegetables, the property has an upper-bound egg-capacity design point of 100 hens (280–310 eggs/hen/year), 100 high-laying Chinese-type ducks (Brown Tsaiya up to ~320 eggs/year), and 100 Muscovies modelled at an upper-bound intensive-line figure of up to ~210 eggs over two reproductive cycles. Costs are anchored by a £270,000 start-up outlay, ongoing leasing of £12,000 per year, self-supplied water, and monthly electricity and feed costs stated in USD and converted to GBP using HMRC’s January 2026 monthly exchange rate. The result is a structured blueprint for a later, journal-grade study, but presented here as a year-one estimate with explicit assumptions and an accounting boundary that excludes orchard yield until measured.
Keywords
Agricultural economics, household provisioning, community food systems, diversified smallholding, protected cropping, hydroponics, integrated pest management, enterprise budgeting, cost accounting, poultry integration, allotment productivity, Diggers Club
1. The claim and the boundary
The year-one claim is narrow by design: a four-acre, mixed system can provision a defined number of families with vegetables as a primary staple stream, while also producing an auxiliary stream of eggs that materially increases household food security and dietary completeness. “Provisioning” in this first-year frame means regular, planned distribution of edible produce to households from the property’s own output, measured on a weekly cadence where possible, and expressed in auditable units: bed area brought into production, cycle counts, harvested units or weights, and household allocations. It does not mean theoretical self-sufficiency across all food groups, nor does it mean a monetised revenue maximisation model. It is a supply-and-cost statement about what this system can deliver when managed as a communal production platform rather than a hobby garden.
Vegetables sit inside the boundary as the principal output because they are the dominant throughput of the polytarp tunnel plots, the hydroponic greens module, and the outdoor beds. Eggs sit inside the boundary as a secondary, separately accounted provisioning stream, treated as a distinct module rather than rolled into vegetable equivalence. This is deliberate: combining eggs into a single “food equivalent” obscures the different constraints, labour rhythms, and input structures that drive poultry versus cropping.
Orchard production is excluded from year-one output totals because the trees are immature and yields are not yet measured; no fruit is credited in the provisioning count. The orchard is still recognised inside the system boundary as a strategic land allocation and an establishment cost that will shape the multi-year economics, but its benefits are deferred until they are observed and recorded. In short: year one counts what is produced and distributed now, assigns costs that are incurred now, and refuses to credit outputs that are not yet real.
2. The physical system in one page: land, tunnel, beds, water, animals
The property is treated as a production platform with a defined footprint, constrained resources, and separable modules that can be costed and audited. The total land area is four acres. Within that footprint, the primary controlled-environment asset is a 1,000 m² polytarp tunnel configured to support intensive, overlapping vegetable cycles. Inside the tunnel, the system is organised around discrete garden plots rather than an undifferentiated open floor. Each full plot is 40 m² of productive bed space, and there are eighteen full plots. In addition, there are two partial plots of 20 m² each. This internal geometry matters because walkways, access lines, and working space are not treated as productive bed area; they are necessary infrastructure, but they do not produce harvest. The result is that “productive area” is defined explicitly as planted bed space, not the gross tunnel footprint.
Outside the tunnel, there is a further 200 m² of outdoor vegetable production area, used to extend total throughput and to carry crops that either benefit from open-field conditions or are allocated outside to balance tunnel space against seasonal needs. The outdoor beds are treated as part of the same provisioning system rather than a separate garden, because their outputs are intended to be pooled into the same household distribution model and assessed under the same accounting boundary.
Water is self-supplied via a well. This changes the operating economics and the constraints simultaneously. It removes routine retail water charges from the cost base, but it does not create “free water” in an engineering sense: reliability, pumping energy, filtration and maintenance, and seasonal variability become the binding considerations. The well supply therefore functions as an input with a different cost profile—more capital and maintenance weighted, less variable-billing weighted—and it supports both the protected cropping environment and the hydroponic greens module that draws well water within a 3 m × 3 m production unit.
Animals are integrated as functional components of the production logic, not ornamental additions. Quail and bantam chickens are used within the polytarp tunnel as a pest-control and sanitation mechanism, reducing pest pressure and contributing manure value. Beyond the tunnel, mixed poultry capacity is designed as an upper-bound stocking potential—ducks, Muscovies, and chickens—providing eggs as a secondary provisioning stream while also contributing to nutrient cycling and biological control. The animals are treated as modules with their own input requirements and labour rhythms, whose value is quantified through eggs and avoided losses rather than assumed as a generic “farm benefit.”
3. Translating bed area into “families fed”: the Diggers benchmark
The translation from planted area to “families fed” is anchored to a simple, published benchmark that is widely understood by gardeners and readily interpretable by non-specialist readers. The Diggers Club expresses productive capacity as a rule-of-thumb: a single 10 m² garden bed can feed one person, and 40 m² of productive bed space can feed a family of four. This is not treated as a law of nature; it is treated as a practical conversion heuristic that allows a first-year estimate to be stated in household terms without pretending precision that the current data do not yet support.
Applying that benchmark to the measured productive bed areas produces an immediate capacity estimate. Inside the polytarp tunnel, the system contains eighteen full plots of 40 m² each, giving 18 × 40 m² = 720 m² of productive bed space. In addition, there are two partial plots of 20 m² each, giving 2 × 20 m² = 40 m². The tunnel therefore contains 720 m² + 40 m² = 760 m² of productive bed area. Under the Diggers conversion, 760 m² divided by 40 m² per family yields 19 families of four. Expressed in people rather than households, the same calculation is 760 m² divided by 10 m² per person, which yields 76 people.
The outdoor vegetable area adds a further 200 m² of productive bed space. Adding this to the tunnel’s 760 m² gives a total productive vegetable bed area of 960 m² for year one. Under the same Diggers benchmark, 960 m² divided by 40 m² per family yields 24 families of four. In per-person terms, 960 m² divided by 10 m² per person yields 96 people. These are the headline provisioning capacities that frame the communal model in a way that is both intelligible and explicitly grounded in a stated conversion rule.
This estimate is a disciplined heuristic, not a claim of measured kilograms of produce, and it is not a guarantee of year-round uniformity without gaps. Its purpose is to define what the system is attempting to do and to give an order-of-magnitude capacity statement that can be audited later. The moment harvest weights, crop losses, quality rejection rates, and household distribution logs are recorded on a weekly cadence, the heuristic can be stress-tested: the estimate either holds under measured output and wastage, or it is revised downward (or upward) with explicit empirical justification.
4. Year-one cropping rhythm: weekly planting, overlapping beds, and why protected cropping matters
The year-one operating model is built around cadence rather than sporadic bursts of planting followed by feast-and-famine harvesting. The core discipline is weekly establishment. Each week, a new 40 m² plot is planted, and in weeks where labour and seedling readiness allow it, multiple plots are planted to maintain momentum or to catch up after weather disruption, pest pressure, or a crop failure. This is not a decorative scheduling preference; it is the basic mechanism by which a small property turns into a provisioning platform capable of supplying several households without collapsing under labour peaks. A weekly planting rhythm converts the biology of crop growth into a predictable pipeline: planting becomes a repeated task with a stable time budget, and harvesting becomes a staggered flow rather than an occasional, overwhelming event.
Overlapping successions follow from that discipline. Instead of treating each plot as a single crop with a single harvest moment, the plots are managed as time-staggered production units. Fast greens, mid-cycle vegetables, and longer-duration crops are sequenced so that the tunnel always has some beds in establishment, some beds at peak production, and some beds in clearance and replant. The overlap is the point. A communal model cannot rely on a single seasonal glut to justify “feeding families” because families eat weekly. The provisioning claim therefore rests on continuity: the system must generate a stable stream of edible output across months, even if the composition changes with season and preference. The weekly rhythm is the simplest way to operationalise that continuity without turning the farm into a spreadsheet fantasy. It creates a production calendar that households can plan around, and it creates a duty roster that is repeatable rather than negotiated anew every fortnight.
Protected cropping inside the polytarp tunnel is what makes this cadence viable under real conditions. Cooling and optimisation shift the probability distribution of outcomes. Open-field beds are exposed to rainfall extremes, heat spikes, wind, and sudden pest explosions; any one of those can compress a week’s labour into a frantic salvage operation or wipe out a succession that the communal schedule assumed would be there. The tunnel does not eliminate risk, but it reduces volatility by controlling temperature peaks, dampening weather shocks, and enabling more consistent germination and early growth. That consistency matters most for greens, where small changes in temperature and moisture can swing a crop from marketable to bolted, bitter, or lost. In economic terms, the tunnel converts a portion of production from high-variance output to lower-variance output. Lower variance is not a vague comfort; it is what allows the communal model to function as an allocation system. When yield is less erratic, labour can be planned, household shares can be honoured more consistently, and the entire structure becomes auditable rather than anecdotal.
5. The hydroponic greens engine: 3 m × 3 m well-watered module, vertical surfaces, and cycle throughput
The hydroponic greens engine is the highest-frequency production unit in the system, and it is treated as such: a compact footprint designed to turn water, nutrients, and labour into a reliable stream of edible greens with short cycle times. The core module occupies a 3 m × 3 m area, supplied from well water, and planted on standard lettuce and rocket rhythms. In year-one terms, the decisive attribute is not the footprint itself but the throughput it supports. Under the operating practice described, the module produces around 1,200 units of mixed greens per cycle. The word “unit” is used deliberately as a harvest-count measure at this preliminary stage, because the present aim is to map system capacity and cadence before moving to a weighed-output dataset; later measurement will translate these units into grams, rejected fractions, and household portions.
Vertical growing surfaces inside the protected environment amplify this throughput without expanding the ground footprint. They create a second layer of productive area that is not in direct competition with bed space used for heavier vegetables. Economically, this matters because the constraint in a communal provisioning system is rarely “total land exists” and almost always “usable growing area that can be managed predictably with finite labour.” Vertical surfaces exploit the tunnel’s enclosed volume, increasing output per square metre of floor area while keeping the work within a controlled microclimate where the probability of crop continuity is higher.
The well-water supply changes the cost and failure profile of the hydroponic subsystem. It reduces variable billing exposure, but it makes reliability, pumping, and maintenance the key constraints. A hydroponic greens engine cannot tolerate the same degree of interruption that an outdoor bed can absorb; a short outage can move a crop from healthy to stressed, and stress translates into lower quality, higher loss, and slower regrowth. In the preliminary accounting boundary, the system therefore treats the well-watered hydroponic engine as a managed, monitored production unit rather than a passive installation. The point is to preserve its role as a stabiliser: when outdoor beds are between harvests, when a tunnel bed is in establishment, or when weather knocks back open-field yield, the greens engine provides a dependable flow of edible material that maintains household baskets and protects the weekly cadence.
The cycle throughput figure—around 1,200 greens per cycle—is not used to overstate year-one conclusions, but to identify the system’s buffering capacity. It is a mechanism for smoothing supply, supporting variety, and making the communal distribution model credible. The later empirical upgrade is obvious and built-in: record cycle length, planting density, harvest weights, and rejection rates, then tie those to household allocations. For now, the hydroponic module is presented as the high-frequency engine that turns controlled conditions into continuity.
6. Integrated pest control and poultry as production infrastructure, not ornament
Integrated pest control in this system is treated as production infrastructure, not as an aesthetic preference or an “organic” badge. Quail and bantam chickens are deployed within the polytarp tunnel as working components of the cropping system, performing two linked functions: biological control and sanitation. They interrupt pest life cycles by foraging for insects and larvae, and they reduce the accumulation of plant debris and fallen material that otherwise becomes habitat for pests and disease vectors. In a controlled environment where crops are planted and harvested on a tight cadence, the cost of a pest outbreak is not merely a damaged bed; it is a break in continuity that propagates into household shortfalls, emergency replanting, and labour spikes. The birds are therefore framed as a risk-management tool aimed at protecting expected output under a weekly succession model.
The economic value is expressed in the only terms that matter for a provisioning platform: reduced crop loss and reduced interventions. Reduced crop loss is the avoided cost of losing harvestable produce, including the waste of inputs already committed—seedlings, nutrients, and labour. Reduced interventions means fewer reactive actions such as manual pest removal, repeated inspection rounds, or additional control measures that consume time and disrupt the planting rhythm. This is not presented as a vague claim that birds “help”; it is presented as a functional substitution, where the birds’ routine activity replaces a portion of human labour and reduces the probability of yield volatility. In a communal structure, that reduction in volatility is as valuable as absolute yield, because it sustains trust in allocations and stabilises the duty roster.
At the same time, the system is explicit about constraints. Full stocking levels and continuous deployment inside the tunnel are limited by practical boundaries: the area that can be safely allocated without damaging plants, the time available for husbandry tasks, and the need to prevent the animals from becoming a source of stress, contamination, or mechanical damage to crops. For that reason, bird numbers and stocking intensity inside the tunnel are treated as operational choices that vary with crop stage and season, not as a constant state. Any headline poultry capacity figures are framed as upper bounds—what the property can support under standard feeding—rather than as a claim that all animals are always present at maximum stocking while the tunnel simultaneously runs at peak cropping intensity. The point of the birds is to increase reliability of output, not to inflate theoretical capacity through simultaneous maxing of every subsystem.
7. Egg provisioning as an “upper-bound capacity” module
Eggs are treated as a separate provisioning module with its own capacity ceiling, labour rhythm, and input structure. That separation is deliberate. Vegetables are a bed-area and cycle-throughput system; eggs are a bird-number and laying-rate system. Blending them into a single “food equivalent” would hide the constraint that poultry output is shaped by genetics, seasonality, broodiness, and feeding, whereas vegetable output is shaped by planting cadence, protected climate, and bed turnover. The year-one objective is therefore to lock the arithmetic of an upper bound—what the property can support under standard feeding and high-layer assumptions—without pretending that this ceiling is the same thing as realised production under day-to-day yarded management.
Start with hens. The property is assumed capable of supporting 100 chickens, and the laying rate is taken from the stated operational expectation: 280–310 eggs per hen per year. The arithmetic is direct. At 100 hens, the annual output range is 100 × 280 = 28,000 eggs at the lower end and 100 × 310 = 31,000 eggs at the upper end. This is the chicken egg ceiling used in the year-one model, recognising that moulting and seasonal conditions will alter the month-to-month profile even if the annual total sits within the range.
For “standard ducks,” the chosen ceiling assumption is the highest Chinese-type layer being used as the reference line for research design. Brown Tsaiya ducks are documented as capable of laying up to roughly 320 eggs per year under high-performance conditions. Using that as an upper-bound genetic and management benchmark, 100 ducks imply 100 × 320 = 32,000 eggs per year. This number is not a prediction of routine backyard performance; it is a deliberate ceiling figure for the research design, setting the high end of what “standard ducks” can deliver when the genetics and management are at the upper end of the distribution.
For Muscovies, the ceiling is set using a peer-reviewed intensive-line figure reported as up to roughly 210 eggs in total across two reproductive cycles. Under the two-cycle framing, 100 Muscovies imply 100 × 210 = 21,000 eggs per year as the upper-bound design figure. This reflects the reality that Muscovy output is often season-structured rather than uniform and that broodiness and reproduction patterns can materially affect realised yield.
Combining the three ceilings yields an annual egg-capacity envelope. At the low end of the hen range: 28,000 (hens) + 32,000 (high-layer ducks) + 21,000 (Muscovies) = 81,000 eggs per year. At the high end of the hen range: 31,000 + 32,000 + 21,000 = 84,000 eggs per year. This combined ceiling provides a useful scale for household provisioning discussions. If a family of four consumes, for example, 24 eggs per week (six eggs per person per week), that is 1,248 eggs per family per year. On that consumption frame, 81,000–84,000 eggs corresponds to approximately 65–67 family-years of that ration. If instead the communal allocation target is 12 eggs per family per week, that is 624 eggs per family per year, and the same capacity corresponds to roughly 130–135 family-years of allocation. The point of these translations is not to assert a single “correct” consumption model, but to show how egg output can be rationed as a stable supplement across many households even when vegetables remain the primary provisioning claim.The cost frame for year one is intentionally conservative in structure and narrow in what it claims. The objective at this stage is not to produce a fully specified farm enterprise budget with every imputed cost and shadow price, but to lock the cash outlays that are already known and to state clearly what is inside and outside the accounting boundary. In year one, the estimate uses the stated start-up capital expenditure, the stated leasing payments, and the stated recurring monthly operating costs for electricity and feed. Labour is not yet priced because the time-use log has not been normalised into a stable weekly pattern across seasons. Depreciation and replacement schedules are also deferred until a complete asset register is recorded with lifetimes and maintenance profiles, so that annualised capital cost is not guessed.
Start-up capital expenditure is treated as a single capex figure of £270,000. In year-one reporting, this is not folded into a per-family cost as though it were consumed instantly. It is recorded as establishment capital that enables the productive system to exist, and it becomes annualised only in the later journal-grade version once asset lives are specified (for example, tunnel structure, pumps, fittings, cooling equipment, fencing, and housing). For the preliminary estimate, it is presented as the entry cost of building the platform rather than an annual operating cost.
Fixed annual recurring cost is dominated by leasing. The leasing arrangement is £6,000 paid twice per year, producing a fixed annual cash outlay of £12,000. This is treated as a non-negotiable carrying cost of the site and therefore sits inside the year-one boundary regardless of whether output is high or low in a particular month.
Variable monthly operating costs are stated in USD: electricity at $100 per month and feed at $88 per month, totalling $188 per month. For consistency in the economic narrative, these are converted into pounds using the stated HMRC January 2026 monthly exchange rate of 1.3329 USD per £1. That conversion implies approximately $1 ≈ £0.7502. On that basis, $188 per month corresponds to roughly £141 per month, which is approximately £1,692 per year when rounded.
Combining the fixed annual leasing (£12,000) with the converted annual electricity and feed (£1,692) yields stated recurring cash costs of approximately £13,692 per year, before any additional variable inputs not yet priced (seed, nutrients, bedding, veterinary costs, repairs, packaging, or fuel), and before labour, depreciation, or contingencies are introduced. Water is self-supplied, so there is no metered water charge in the cash cost line; the operating cost of water is instead embedded in pumping and system operation, which is already captured, at least partially, through the electricity figure, with filtration and maintenance costs to be added later once logged.
This remains capacity arithmetic. It sets the ceiling. The empirical work that follows is to record realised laying under free-range yarding conditions, including seasonal profiles, broodiness effects, feed conversion realities, and mortality, then to compare observed output to this ceiling to establish a defensible utilisation rate for the system in year one.
8. Costs: start-up, fixed annual costs, variable monthly costs
The cost frame for year one is intentionally conservative in structure and narrow in what it claims. The objective at this stage is not to produce a fully specified farm enterprise budget with every imputed cost and shadow price, but to lock the cash outlays that are already known and to state clearly what is inside and outside the accounting boundary. In year one, the estimate uses the stated start-up capital expenditure, the stated leasing payments, and the stated recurring monthly operating costs for electricity and feed. Labour is not yet priced because the time-use log has not been normalised into a stable weekly pattern across seasons. Depreciation and replacement schedules are also deferred until a complete asset register is recorded with lifetimes and maintenance profiles, so that annualised capital cost is not guessed.
Start-up capital expenditure is treated as a single capex figure of £270,000. In year-one reporting, this is not folded into a per-family cost as though it were consumed instantly. It is recorded as establishment capital that enables the productive system to exist, and it becomes annualised only in the later journal-grade version once asset lives are specified (for example, tunnel structure, pumps, fittings, cooling equipment, fencing, and housing). For the preliminary estimate, it is presented as the entry cost of building the platform rather than an annual operating cost.
Fixed annual recurring cost is dominated by leasing. The leasing arrangement is £6,000 paid twice per year, producing a fixed annual cash outlay of £12,000. This is treated as a non-negotiable carrying cost of the site and therefore sits inside the year-one boundary regardless of whether output is high or low in a particular month.
Variable monthly operating costs are stated in USD: electricity at $100 per month and feed at $88 per month, totalling $188 per month. For consistency in the economic narrative, these are converted into pounds using the stated HMRC January 2026 monthly exchange rate of 1.3329 USD per £1. That conversion implies approximately $1 ≈ £0.7502. On that basis, $188 per month corresponds to roughly £141 per month, which is approximately £1,692 per year when rounded.
Combining the fixed annual leasing (£12,000) with the converted annual electricity and feed (£1,692) yields stated recurring cash costs of approximately £13,692 per year, before any additional variable inputs not yet priced (seed, nutrients, bedding, veterinary costs, repairs, packaging, or fuel), and before labour, depreciation, or contingencies are introduced. Water is self-supplied, so there is no metered water charge in the cash cost line; the operating cost of water is instead embedded in pumping and system operation, which is already captured, at least partially, through the electricity figure, with filtration and maintenance costs to be added later once logged.
9. What “communal” changes economically: allocation rules, labour discipline, and auditability
“Communal” is not treated here as a sentiment or a vague social good. It is treated as a governance technology that converts biological yield into household welfare by imposing rules on allocation, labour, and timing. A four-acre property can produce a great deal in bursts and still fail to provision families if the product arrives irregularly, if labour is concentrated on a few people, or if distribution becomes a dispute rather than a process. The communal structure is the mechanism that prevents that failure. Weekly bed planting creates the production pipeline; harvest allocation converts that pipeline into predictable household baskets; duty rosters ensure the pipeline does not collapse under uneven contribution. The economic point is simple: governance reduces variance. Reduced variance raises the realised value of output, because a smaller but reliable weekly basket can be more welfare-improving than a larger, erratic glut followed by gaps.
Allocation rules are the first economic lever. If the system is provisioning 24 families on a bed-area benchmark, the distribution method must specify whether households receive equal shares, shares proportional to household size, or shares linked to labour contribution. Each choice has consequences. Equal shares are administratively simple but may be contested by larger households. Household-size shares align output with consumption needs but can create free-rider problems unless labour is also structured. Contribution-linked shares can protect fairness perceptions but risk turning the system into a ledger contest rather than a provisioning platform. Whatever rule is chosen, it must be explicit, stable, and enforceable without endless negotiation. That stability is part of the production function, because it determines whether people show up reliably to plant, harvest, clean, and reset beds on the weekly cadence.
Labour discipline is the second lever. The system requires a duty roster that assigns repeating tasks—planting, harvesting, tunnel checks, hydroponic monitoring, animal husbandry, composting, cleaning, repairs—so that no single person becomes the bottleneck. The roster is not merely organisational; it is economic, because it reduces the transaction cost of coordinating labour and prevents the collapse of planting cadence that would immediately translate into future output gaps. In a communal model, labour is the scarce factor that determines whether theoretical capacity becomes realised provisioning.
Auditability is the third lever, and it is what makes the claims credible. A communal model can be described persuasively and still be empty if it cannot be verified. The minimum record set is plain: harvest logs by bed and by week (counts and, later, weights); distribution logs recording what each household actually received; labour logs capturing hours by task and by person; and input receipts for seeds, nutrients, feed, electricity, replacements, and repairs. With those records, the provisioning claim becomes testable, disputes become solvable, and the system can be analysed as an economic structure rather than as a narrative.
10. Household economics: annual cost per family, labour contribution, and avoided grocery expenditure (vegetables and eggs only)
The year-one household economics is framed as a two-part balance sheet: a cash ledger that keeps the platform operating and a household provisioning ledger that measures what the platform replaces in ordinary grocery purchasing. On the cost side, the annual recurring cash outlay already stated for the system is £13,692, comprising £12,000 per annum in lease payments and approximately £1,692 per annum for electricity and feed on the stated conversion basis. Allocated across a 24-family provisioning scale, this yields a direct operating cash burden of approximately £570.50 per family per year, or approximately £47.50 per family per month. This figure is deliberately narrow: it is the year-one cash requirement to keep the site functioning under the present boundary, not a full economic cost that would also include seeds, nutrients, repairs, veterinary inputs, consumables, replacement parts, and capital annualisation.
Labour is treated as an explicit, non-trivial household contribution rather than an implicit assumption. The communal discipline is specified as 12 hours per week per family. Annualised, this is 624 labour-hours per family-year. At a 24-family scale, the communal labour pool becomes 14,976 hours per year. In year one, that labour is not treated as a cash expense paid out by the project; it is an in-kind contribution that substitutes for hired labour. Its economic treatment is therefore bifurcated. First, it is recorded as a quantity constraint and a governance requirement: the system’s realised output and continuity are conditional on those hours actually being supplied and properly rostered. Secondly, it can be monetised later as an imputed cost using a stated shadow wage rate, enabling a formal comparison between a household’s “true” resource cost and their avoided grocery expenditure. Without a declared shadow wage, the analysis remains valid in its primary form: families exchange time for food, and the relevant question becomes whether the food received is worth more than the groceries displaced, given the time commitment and the cash contribution.
On the benefit side, the appropriate comparator is not total household food spend but the retail value of an equivalent basket of vegetables and eggs. The vegetable component is tied to the provisioning capacity benchmark (24 families) and then operationalised through a defined weekly allocation rule once distributions are recorded. Eggs can be treated in the same way, with the allocation rule calibrated against the capacity envelope already established for high-layer assumptions. Under this structure, each household’s annual avoided expenditure is computed as the sum over weeks of the retail value of the vegetable share plus the retail value of the egg share, using observed quantities and contemporaneous local prices. Net benefit per family is then expressed in two parallel metrics: a cash metric (annual avoided grocery spend minus £570.50 annual cash contribution) and a full resource metric (annual avoided grocery spend minus £570.50 minus the imputed value of 624 hours). This yields a defensible, auditable family-level economic statement that does not require inflated claims about meat, orchard yield, or unmeasured outputs in year one.
11. The research roadmap: turning a year-one estimate into a journal paper
The year-one estimate is intentionally framed as a disciplined preliminary: it translates bed area and system configuration into a household provisioning capacity using an explicit heuristic, and it separates vegetables and eggs into auditable modules. The journal-grade upgrade is therefore not conceptual; it is empirical. The minimal dataset required is the smallest set of records that converts “capacity as geometry and cadence” into “capacity as observed output and realised allocation,” with enough structure to support replicability and external review.
The foundation is a bed-level harvest dataset. For each bed (or plot) and for each week, the system records harvested quantities in measured weights, with a consistent unit definition and a traceable mapping to crop type. Counts can be retained for operational convenience, but the journal paper requires weights to support nutritional equivalence, waste accounting, and cost per kilogram calculations. Alongside harvested weights, wastage must be recorded as a first-class variable, not an afterthought: losses at harvest, post-harvest rejection, spoilage before distribution, and any crop failure events. These loss records are what allow the paper to distinguish “gross biological yield” from “net edible provisioning,” which is the only quantity that matters for household welfare and economic substitution.
Labour must be logged at the task level rather than as a single undifferentiated hour total. A minimal labour log records hours by household and by task category (planting, harvesting, tunnel maintenance, hydroponic monitoring, animal husbandry, cleaning and sanitation, repairs, and administration/distribution). This enables both governance analysis (who contributes what, and whether the duty roster is sustainable) and economic analysis (shadow wage valuation under alternative wage assumptions). Egg output must be recorded as counts by flock and by week, with separate tallies for hens, the “standard duck” flock, and Muscovies, so that observed laying profiles can be compared against the upper-bound ceiling assumptions and so that seasonality can be quantified rather than asserted.
A cost ledger is the final required pillar. It should capture all recurring expenditures (lease, electricity, feed) but must also add the operational items omitted from the preliminary boundary: seeds and seedlings, nutrients, bedding, veterinary inputs, repairs, replacements, and any consumables tied to protected cropping and hydroponics. Capital items should be listed in an asset register with purchase date and expected life so that annualised capital cost can be introduced transparently rather than guessed. With these inputs and outputs, the paper can compute cost per kilogram of vegetables, cost per dozen eggs, and cost per family basket under alternative labour valuations.
To connect output to the “families fed” claim in an academically acceptable way, a seasonal continuity index is required: the number of weeks in which a full defined basket was delivered to each family, and the distribution of shortfalls by season and cause. This is the metric that tests whether weekly cadence produces weekly welfare rather than periodic gluts. Orchard treatment remains explicit throughout: orchard yield is set to zero until measured; establishment costs and land allocation are recorded now; and the orchard becomes a multi-year extension study that is analysed separately once fruit output is observed, rather than being used to inflate year-one conclusions.